When I came to CDW five years ago, I began using Microsoft Lync (then called Office Communicator) for all my telephone calls. At the time, many experts expressed doubt that Microsoft could support reliable phone calls using a software-centric approach. But similar doubt was leveled at Cisco when they proposed to deliver reliable voice over IP (VoIP) to a market using digital private branch exchanges (PBXs). VoIP technology quickly matured and Cisco proved the critics wrong. Microsoft has now done the same by demonstrating the simplicity and reliability of Lync as a complete enterprise communications system.

Lync’s simplicity is obvious after only a few minutes of use. Interestingly, because Lync’s unique software-centric approach is so straight forward, it often leads some customers to mistakenly believe that Lync can only be used as a softphone. In fact, there are a wide variety of phones and endpoints to use with Lync. But it is true that Lync’s softphone experience is amazing and can translate into dramatic savings for organizations looking to minimize the tremendous cost of new phones as part of a new phone system.

Reliability, on the other hand, is one topic that the critics fought long and hard over. A new feature in Lync Server 2013, called pool pairing, offers even more options for reliability. First, let’s quickly review the two deployment models of Lync Server 2013.

  1. Single Server Pool – Lync Server can be deployed on a single server. This is simliar to many small PBXs or key systems. The one server is as reliable as the platform it’s running on. Server virtualization technology can be used to provide hardware fault tolerance, improving up-time beyond what a small PBX or key system can deliver.
  2. Multi Server Pool – Two or more Lync servers can be configured using load balancers and mirrored SQL. This configuration introduces more cost and complexity but provides a deployment that is highly available (HA). All components are made redundant to eliminate single points of failure. Failover of all services is automatic if a server becomes unavailable for any reason.

But what if a disaster strikes (such as a fire or flood) that takes out an entire data center? Pool pairing is a new feature of Lync Server 2013 that supports disaster recovery scenarios.

Pool pairing allows two server pools to share information in real time about the state of the entire system. The second server pool should be located far enough away so a disaster is unlikely to affect both pools.

In a disaster scenario, Lync clients will automatically connect to a server of the surviving pool using reduced functionality mode. Reduced functionality includes the ability to make and receive phone calls thereby providing automatic failover for voice. Pool pairing allows the Lync administrator to send a command to the surviving pool, instructing it to begin servicing the affected users for all functions. All system functions are restored for all users in a matter of seconds after the command is entered.

The cool thing about pool pairing, that may not be immediately obvious, is that you can use pool pairing between two single servers even if the two servers are right next to each other. So ignoring the disaster recovery benefits, pool pairing can be used to provide automatic failover for voice and ultimately full restoration of all functions, once the administrator sends the command. The full failover is not automatic, like a true HA pool, but it’s a very simple design. It’s easy to add a second server in a single server environment to add meaningful resilience capability with little added cost. Smaller companies can use the pool pairing feature of Lync Server 2013 to leverage sophisticated failover options without the cost of a true HA deployment. Larger organizations can choose to deploy HA pools to allow automatic failover if a server fails as well as manual recovery from the loss of an entire datacenter. Pool pairing is just one reason why millions of users have switched to Lync for all voice communications. I use Lync every day. I love it. Looking back at the last five years, I can’t imagine using a traditional phone system.

2 thoughts on “Microsoft Lync Server 2013: Pool Pairing for Enterprise Voice

  • Jeffrey David says:

    I was a huge fan of Lynch until security became top priority for my business; hence, switched over to RHUB web conferencing appliances for conducting online meetings. It works from behind my firewall.

  • Salman Aslam says:

    Dear Lync Gurus,
    Could you please help me in the below Scenario.
    I am building Lync 2013 Pair Pooling environment in two different Datacenter locations. Before planning, I need certain clarifications on it.
    1 – I have Single Forest & Single Domain Environment (Both the sites have Internet Connectivity).
    2 – If one Site is down, how other site will take over (Manually or Automatically)
    3 – Do I need to have strong enough WAN connection (point to point) between the datacenter?
    4 – If my WAN connection is failed than how both the sites replicate?
    My major concern if the WAN Connection failed between the both Sites than how Lync replicate since its Single Forest & Single Domain Environment.
    Highly appreciated your positive reply.

Comments are closed.